The disparity is staggering. Africa bears one-quarter of the burden of disease around the world yet has barely 3 percent of all health workers. Millions of people across the continent thus suffer needlessly because they cannot obtain medical care from trained personnel. In sub-Saharan Africa, where the crisis is most acute, fully 820,000 additional doctors, nurses, and midwives are needed to provide even the most basic health services. To meet this shortfall, most of the region’s countries would have to increase the size of their health workforce by 140 percent.
Unfortunately, the money to hire, train, and sustain new recruits on this scale isn’t available and likely won’t be in the foreseeable future. Moreover, the classical Western model of health care training, which prevails throughout much of Africa, is capital intensive and time consuming: even if the funds materialized, about 600 additional medical and nursing schools and more than two decades would be needed to close the gap. Another challenge is the relatively low appeal of health care as a career in Africa. Wages for doctors and nurses are low there, and many doctors depart for the better working conditions, job security, and pay available elsewhere. Finally, poor infrastructure and scarce medical supplies—among other factors—limit the productivity of health workers, so the true extent of Africa’s delivery gap could be even larger than it seems.
Against this grim backdrop, addressing Africa’s health workforce crisis seems to be an impossibly distant prospect. It needn’t be. McKinsey’s work in eight African countries1 and recent interviews with more than 40 experts2 suggest the outlines of a threefold approach that could begin ameliorating the problems within the coming decade. First, countries in Africa should consider successful health delivery models used in some other developing countries, where community health workers and other paraprofessionals complement the work of their professional colleagues. Such approaches save money and training time and help make populations healthier. Second, nongovernmental organizations (NGOs) and other development community partners should support such efforts by collaborating with the region’s governments to help improve the productivity of health workers—for example, by working to close knowledge gaps and supporting morale-boosting incentive schemes. Finally, government leaders in sub-Saharan Africa should reconsider the rules that in many places limit the training of medical professionals to the public sector. Creating an environment that encourages broader participation from the private sector, NGOs, and “social enterprises” would increase desperately needed training capacity and help relieve overburdened public facilities.
A continent in crisis
Across sub-Saharan Africa, thousands of people die each day from malaria, tuberculosis, HIV/AIDS, and other preventable and treatable conditions. The dearth of health delivery workers is partly responsible for this tragedy, and approaches based on the status quo won’t meet the region’s health needs quickly or effectively.
A yawning gap
Attempts have been made to quantify the impact of increasing the number of health workers in Africa,3 but determining the precise number that individual countries need is difficult given differences in their health care systems and the varying levels of productivity their health workers have so far achieved. The development community nevertheless believes that 2.5 health workers per 1,000 people is consistent with the maternal- and infant-health targets embodied in the UN’s Millennium Development Goals.4
Against even this modest benchmark, sub-Saharan Africa performs dreadfully (Exhibit 1). A simple extrapolation of the available data shows what’s needed to reach the threshold (Exhibit 2), but little in the region’s history suggests grounds for optimism. Evidence from the World Bank shows that health delivery coverage there has improved little over the past four decades, particularly when compared with countries such as India, where densities of physicians and nurses have increased three- and fivefold, respectively, since 1960. Our research not only confirms that the number of health workers in sub-Saharan Africa is growing more slowly than its population but also suggests that, without major intervention, the absolute number of physicians there could start decreasing by 2009.
Funding major interventions, however, will be challenging, despite the region’s strong economic growth. If health spending in sub-Saharan Africa continues to grow in line with local economies, for example, and money from donors increases slightly faster than it has in the past, nearly $22 billion will be available for spending on health care in 2015. Of that sum, fully 84 percent will be required to maintain today’s median health spending per capita—a scant $20 per person. After adjusting wages for inflation and taking into account the costs of promising (but expensive) new therapies for tuberculosis and HIV/AIDS, a mere $400 million will be left to finance new hiring (Exhibit 3).
The wrong model
A related obstacle to meeting the demand for health workers is sub-Saharan Africa’s classically Western-inspired health delivery model: 93 percent of the region’s health delivery workers are professionals (doctors or nurses) as opposed to paraprofessionals. Systems based on Western archetypes may provide satisfactory care for the predominantly urban patients they reach, but Africa’s poor infrastructure means that many rural patients receive little or no medical attention. Doctors also usually prefer living in towns and cities, so countries such as Equatorial Guinea, Ghana, and Senegal report four times as many urban physicians as rural ones, compared with a global average of 2.5 to 1.
The classical training model, moreover, requires large investments in facilities and years of formal training for students. Our modeling suggests that attempts to close sub-Saharan Africa’s health delivery gap solely with professionally trained doctors and nurses would require about $33 billion for training costs and approximately 300 more medical schools (in addition to the 90 or so there today), as well as a comparable number of new facilities to train nurses. Even if the region increased its training capacity to the world average—a tall order—more than 20 years would be needed to produce enough health professionals to close the gap.
Morale and productivity problems
What’s more, careers in health care aren’t as attractive in sub-Saharan Africa as they are elsewhere. Nurses there, for example, earn half as much money as teachers do, and doctors get significantly less than engineers. Many physicians are therefore tempted to quit medicine or to leave the continent for the better pay, greater job security, and more agreeable working conditions of the developed world. Perversely, the prevalence of Western training in Africa contributes to this exodus, since local doctors are qualified to work elsewhere. Only 50 of the 600 doctors trained in Zambia since independence continue to practice in that country, according to a 2004 report from the Joint Learning Initiative,5 an international consortium of academic institutions and development agencies. More of Malawi’s doctors may be practicing in the British city of Manchester than in Malawi.
In some countries, government hiring freezes and the inconsistent payment of wages contribute to the problem. In Malawi, for instance, vacancies in the public-health industry are as high as 90 percent for doctors and 60 percent for nurses. In Ghana nearly half of all public-nursing positions remain vacant because of hiring freezes.
Such factors dampen morale among health workers. Along with problems such as poor infrastructure and inadequate medical supplies, they also cut productivity. Although comprehensive evidence is scarce, studies from Chad and Tanzania indicate that, in some settings, health workers spend only 50 to 60 percent of their time on productive activities; productivity could potentially improve by 35 percent in Chad and by 26 percent in Tanzania.6 Red tape is another problem: in Tanzania, for example, we found that district medical officers spent more than 25 days during any three-month period writing reports.
Ameliorating the crisis
Solving Africa’s health delivery crisis will require decades and the concerted efforts of the public and private sectors, as well as the development community. Still, we believe that it will be possible to start facing the challenge within the coming decade if the region’s governments address the demand for health workers by building systems based on thoughtful ratios between professional and paraprofessional workers. Governments can’t achieve this alone; the development community and the private sector have important roles to play.
A more appropriate model
The idea of complementing the work of health professionals by making use of substitute medical doctors, community health workers, and other paraprofessionals isn’t new. From the 1950s through the 1970s, for example, China’s four million “barefoot doctors” and rural health aides significantly improved that country’s life expectancy and helped cut death rates. More recently, Brazil and Iran have successfully employed similar schemes. Iran’s paraprofessional health workers (behvarz) visit the homes of underserved populations, for instance, providing vaccinations and monitoring childhood growth. Brazil uses teams of doctors, nurses, dentists, and paraprofessionals to support impoverished areas. Both countries have significantly improved their health outcomes in this way.7
In sub-Saharan Africa a reliance on Western-based models has meant that paraprofessionals now make up only 7 percent of all health workers, compared with about 20 percent in Brazil and Iran. This is a missed opportunity. In Africa and elsewhere, substitute medical doctors, who deliver many services usually reserved for physicians, need only two to three years of training rather than five years in medical school—a savings of up to $48,000 per worker in training costs. In Tanzania the substitute medical doctors, known as assistant medical officers, provide obstetric services, administer anesthesia, undertake minor surgery, and respond to medical emergencies. Substitute medical doctors cost only one-half to two-thirds as much as junior physicians do, and our interviews suggest that they are easier to retain than doctors, since their certification typically isn’t recognized outside their country of operation.
Paraprofessionals are also reliable: one study of 10,000 operations by substitute medical doctors in Mozambique showed impressively low rates of mortality: 0.1 percent for elective and 0.4 percent for emergency surgery. Results for obstetric surgery in Mozambique and elsewhere show little difference between the results achieved by properly trained substitute medical doctors and by surgeons.8
Community health workers, another type of paraprofessional, can also help meet Africa’s health delivery needs. These workers receive much less training (a few hours to several months) than a substitute medical doctor does. Providing relatively simple, yet equally invaluable, services, they support rural villages with a wide variety of basic services, such as hygiene, sanitation, reproductive health, first aid, and, in some cases, vaccinations and rudimentary interventions, including oral rehydration therapy for infants. Community health workers also provide desperately needed preventive care in rural areas and play a triage role—for instance, by recognizing symptoms that require immediate professional attention.
All of this activity supports thinly stretched national health services. In Niger, for example, community health workers provide the initial treatment for up to half of all patients suffering from diarrhea, conjunctivitis, and malaria. In Saradidi, a village in western Kenya, community health workers not only handle all these tasks but also weigh babies and record local births and deaths, thus allowing physicians and nurses to see more patients.
Such results have already persuaded South Africa, Sudan, and Zambia to plan increases in their paraprofessional workforce; likewise, by 2010 Ethiopia intends to deploy 30,000 community health workers and 5,000 substitute medical doctors trained in hygiene, sanitation, family health services, the prevention and control of disease, health education, and communication. More countries should follow suit. Indeed, our modeling of two illustrative scenarios—one based on the experiences of Brazil and Iran, the other requiring no additional professional training capacity beyond today’s levels—suggests that achieving the World Health Organization’s (WHO) recommended minimum threshold for health workers partly with paraprofessionals would be significantly more cost effective than traditional approaches (Exhibit 4). Paraprofessionals help save lives quickly because they are faster to train than doctors and nurses and make those professionals more productive.
Focus on productivity
Since circumstances differ widely across Africa, just adding more health workers—professional or otherwise—won’t be enough. Countries must implement quality- and productivity-improvement measures at the same time to make the best use of a growing health workforce. The development community has a crucial role to play here, starting by helping African governments to fill the knowledge gaps that lower productivity.
Most countries in sub-Saharan Africa lack reliable measurements of the resources they devote to public health and the results they achieve—not least the state of the rural population’s health. They therefore have difficulty allocating resources efficiently or applying traditional population health measures, let alone new ones. Indeed, few countries in sub-Saharan Africa even know precisely how many health workers they have and where those workers are distributed.9 More than 40 percent of these countries lack data on how many community health workers operate within their borders, for instance. NGOs can help fill such information gaps. They should start by teaming up with the Global Health Workforce Alliance (GHWA) and other multilateral groups to help answer, for example, basic questions about the distribution of human resources at the district level and the availability of infrastructure and medical supplies in rural areas. Current efforts to elicit this information should be expanded.
WHO, which, for example, supports the use of community volunteers to help fight river blindness in the Democratic Republic of the Congo,10 could help by adding information on paraprofessionals to the field data it already collects through NGOs and other groups. Such information—say, the number of community health workers in a district and the average number of patients they see in a quarter or the number of days a substitute medical officer spends in the field—could help NGOs and national governments to improve the allocation of resources, much as countries now use existing field data on professional health workers. Kenya’s Ministry of Health, for instance, recently conducted a mapping exercise to collect basic information, such as where public-health personnel work and how much training they have. Its findings should allow it to optimize their deployment. New information on paraprofessionals would help governments to allocate resources more efficiently, to set budgets, and eventually to establish universal training guidelines.
NGOs and other organizations in the development community should also support Africa’s governments in tailoring incentive schemes to optimize the productivity and morale of health workers. India’s Accredited Social Health Activist (ASHA) program, for example, raised retention levels of community health workers by instituting a fee-for-service honorarium of $20 to $40 a month; in Bangladesh community health workers called shasta shabikas earn wages based on whether their patients complete treatment cycles.
Nonmonetary incentives are also promising. Our work in Tanzania suggests that paraprofessional health workers, like their professional counterparts, place a high value on access to tools and technologies. Providing new training opportunities or products, from soap to mobile phones,11 that help workers do their jobs better can therefore have a big impact. In our experience, these incentives should reward the performance both of individuals and teams and need not exceed about 20 percent of a worker’s base compensation to be effective. Other incentives are free. In India, for example, our interviews suggest that publicly recognizing high-performing paraprofessionals within their communities serves as a powerful motivation tool.
Beyond the public sector
Since paraprofessional health workers should complement, not replace, medical professionals, Africa must also bolster the ranks of its doctors and nurses. There are many competing priorities for scarce public funds, so the private sector, NGOs, and other groups with a social mission should help—for example, by creating private alternatives to oversubscribed public training facilities. Our interviews in Ghana, for example, suggest that 60 percent of its qualified nursing applicants are rejected because of insufficient capacity.
Most governments in the region, however, limit private involvement in medical training. Among other things, they have a historical sense of unease about the idea that the private sector should play a role in medicine (in particular, about private delivery) and fear that the quality of training would be lower at private institutions than it is at public ones.
These are legitimate concerns. Still, with appropriate government oversight, private training programs could help. In Senegal, for instance, 60 percent of the nurses and 45 percent of the midwives are educated privately. One school, L’Institut Santé Services (ISS), trains about 100 nurses a year and boasts student exam scores comparable to those of public schools. The demand for placement at its Dakar campus is two times capacity, and 85 percent of its graduates go on to work in Senegal’s public-health system. In Tanzania the nonprofit institution Hubert Kairuki Memorial University, in Dar es Salaam, offers a number of professional medical degrees and has increased its enrollment sixfold, to 200 students, since its founding in 1997. In Kenya the African Medical & Research Foundation (AMREF), an NGO, balances classroom and remote instruction (via the Web) to train nurses. Since 2005 it has expanded from 4 sites and 145 students to 127 and 4,500, respectively—20 percent of Kenya’s nursing students. Moreover, training people closer to their homes may increase retention levels; our interviews suggest that up to 70 percent of AMREF’s rural graduates remain in their localities after graduation. Remote-learning models also hold promise for Nigeria and Tanzania, where candidates are highly dispersed yet can reach the necessary infrastructure (in commercial centers or major hospitals, for example).
African governments that support such efforts should consider creating financing programs to help qualified students pay for private instruction, which for nurses can be up to 50 percent more expensive than its public counterpart. Few governments do so today, mainly because more than enough qualified candidates can pay. As training capacity grows, however, so will the pool of qualified students requiring financial aid. Funding for loans could come from a combination of public, commercial, and philanthropic sources (as it does in the United States). Subsidized loans might require a postgraduate service commitment.
To help maximize the return on these educational investments, African governments might look to countries such as Thailand (see sidebar, “Doctors where they’re needed”), which helps to ensure high-quality instruction at private medical schools by requiring their graduates to pass a licensing exam not demanded of public-school graduates. Some private hospitals in Thailand optimize staffing levels by offering scholarships to graduate nursing students in exchange for a few years of service.
The health workforce crisis in sub-Saharan Africa presents daunting challenges. To begin tackling them, the region’s governments should build health systems with a thoughtful ratio between professional and paraprofessional workers and cooperate with the development community and the private sector to train more medical professionals and to raise their productivity. 
About the Authors
Michael Conway is a director in McKinsey’s Philadelphia office, Srishti Gupta is a consultant in the Boston office, and Kamiar Khajavi is a consultant in the Washington, DC, office.
The authors wish to thank Kate Barrett, Aida Causevic, and Todd Johnson for their contributions to this article.
Notes